Our visual experience is of a rich, detailed world, yet decades of experiments in visual perception demonstrate that we can attend to only a small portion of it at any one time. Traditionally, theories of perception have posited a single processing stream that includes a massively parallel processing stage followed, after a selective bottleneck, by a stage limited to the processing of one or, perhaps, a few objects at a time. While such theories have successfully explained a variety of experimental phenomena, a series of recent studies have called into question the existence of this single processing stream. However, these new findings are controversial. The goal of the present project is to rigorously test these new observations and to extend them if they prove reliable. The results will have far-reaching implications for our understanding of visual processing, potentially reconciling the conflict between experimental results and visual experience.